acquire-codebase-knowledge

Generate seven evidence-based documentation files describing an existing codebase's stack, structure, and architecture.

Updated Sep 10, 2026
One-click install
npx skills add https://github.com/serpro-workshop-fortaleza/datacorp-sifap-modernization-team-kit --skill acquire-codebase-knowledge-serpro-workshop-fortaleza
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: acquire-codebase-knowledge
Source: https://github.com/serpro-workshop-fortaleza/datacorp-sifap-modernization-team-kit/tree/main/.github/skills/acquire-codebase-knowledge
Command: npx skills add https://github.com/serpro-workshop-fortaleza/datacorp-sifap-modernization-team-kit --skill acquire-codebase-knowledge-serpro-workshop-fortaleza

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? New engineers and AI assistants lack verified, structured knowledge about an unfamiliar repository, leading to wrong assumptions about its stack, architecture, and conventions. This Skill produces seven evidence-backed documents in docs/codebase/ so anyone can become productive on the project quickly. ## Core Features & Use Cases - Automated repository scan: A Python script collects directory trees, manifests, entry points, lint configs, TODOs, git churn, CI/CD, containers, and security configs. - Seven structured templates: STACK, STRUCTURE, ARCHITECTURE, CONVENTIONS, INTEGRATIONS, TESTING, and CONCERNS documents filled only with verifiable claims, using [TODO] and [ASK USER] markers for unknowns. - Validation loop: Every claim must trace to a file, config, or terminal output before the documentation is accepted. - Use Case: A developer joining a legacy modernization project asks the AI to map the repository, and receives a complete, evidence-cited documentation set covering technology stack, data flow, integrations, and technical debt. ## Quick Start Ask the AI to map this codebase and document its architecture, stack, and conventions in docs/codebase/.

Frequently Asked Questions about acquire-codebase-knowledge

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I document an existing codebase automatically?▼

Run the scan script from the repository root, then fill seven templates (STACK, STRUCTURE, ARCHITECTURE, CONVENTIONS, INTEGRATIONS, TESTING, CONCERNS) using only claims verifiable in files or terminal output. Unknowns are marked [TODO] or [ASK USER].

How to detect a project's technology stack from its files?▼

Read manifest files like package.json, go.mod, pyproject.toml, pom.xml, or Cargo.toml to identify the ecosystem, runtime version, and dependencies. The stack-detection reference maps each manifest and dependency to its framework.

Does the scan script work on Windows and monorepos?▼

Yes, scan.py is cross-platform Python 3.8+ with no external dependencies. It detects monorepo signals like pnpm-workspace.yaml, nx.json, turbo.json, and package.json workspaces, and each subpackage should be mapped separately.

What are the limitations of automated codebase documentation?▼

The skill only documents what is verifiable in source files; it never infers intent. Team decisions, planned features, and ambiguous architecture choices must be marked [ASK USER] and resolved with the team rather than guessed.

Why should generated directories be excluded from codebase documentation?▼

Directories like dist/, build/, .next/, and __pycache__/ contain compiled artifacts, not source conventions. Documenting them produces false claims about naming and structure, so the scan excludes them by default.